Nitrogen management under increased atmospheric CO2 concentration in cucumber (Cucumis sativus L.): ameliorating environmental impacts of fertilization

In the last years, the atmospheric CO2 concentration has increased significantly, and this increase can cause changes in various physiological and biochemical processes of plants. However, the response of plants to elevated CO2 concentration (e[CO2]) will be different depending on the nitrogen form available and the plant species. Therefore, hydroponic trials on cucumber plants, with two CO2 concentrations (400 and 1000 ppm) and two nitrogen sources (NO3−/NH4+; 100/0 and 90/10), were conducted. Physiological parameters—such as gas exchange, GS, GOGAT and GDH activities, cation composition, soluble sugar and starch content- were measured. The results showed that when plants were grown with NH4+ and e[CO2], parameters such as photosynthesis rate (ACO2), instantaneous water use efficiency (WUEi), the content of NH4+, Ca2+ and Mg2+, and the concentration of starch, were higher than in control plants (irrigated with nitrate as sole nitrogen source and ambient CO2 concentration (a[CO2])). Furthermore, an improvement in N assimilation was observed when the GS/GOGAT pathway was enhanced under these conditions (NH4+ and e[CO2]). Thus, our results contribute to the reduction of the negative environmental impacts of the use of nitrogen fertilizers on this crop, both by reducing nitrogen leakage (eutrophication) and greenhouse gas emissions.

appropriate ratio of inorganic nitrate under future climate scenarios. The nitrogen fertilization of crops is one of main causes of environmental contamination worldwide, through nitrate leaching and as a net contributor to greenhouse gas emission 7,8 . Cucumber (Cucumis sativus L.) is one of the most-cultivated vegetables in the world due to its economic and nutritional benefits 5,9 . However, although many cucumber growth parameters have been studied, such as photosynthesis 10 , nitrogen metabolism 11 , fruit quality 12 , or water use efficiency 13 , the combined effects of CO 2 and different N forms (NO 3 − and NH 4 + ) have not been studied. Therefore, this study is the first attempt at understanding how N forms and e[CO 2 ] interact in cucumber plants in a climate change scenario. To stimulate the physiological mechanisms affected by these two factors (N forms and e[CO 2 ]), cucumber plant seedlings were exposed to different N inputs and e[CO 2 ] in a controlled environment. The responses of plants were assessed by measuring the net CO 2 assimilation, internal CO 2 concentration, instantaneous water use efficiency, cation concentration, nitrogen-metabolizing enzymes, starch, and soluble sugars.

Results
Gas exchange. The data showed that the treatment with NH 4 + (90/10) increased the A CO2 at both CO 2 concentrations (Fig. 1A). The A CO2 throughout the experiment slightly decreased at both CO 2 concentrations. Curiously, the decrease was more pronounced when the plants were irrigated with the 100/0 treatment at both CO 2 x* x* ; and (C) instantaneous wateruse efficiency (WUEi). Data are means ± SE of six plants. * denotes significant differences (P < 0.05) between plants in different CO 2 treatments, with the same nitrogen treatment; X denotes significant differences (P < 0.05) between nitrogen treatments for the same CO 2 treatment. Soluble sugars and starch. The soluble sugars and starch contents were affected by N form ( Fig. 2A    www.nature.com/scientificreports/ Nitrogen-metabolizing enzymes. The behaviors of these 3 enzymes (GS, GOGAT, and GDH) against the N source were dependent on the environmental conditions in which the plants were grown (Fig. 3). Plants grown under a[CO 2 ] and irrigated with the 90/10 treatment, suffered a significant reduction in GOGAT activity (62%), and an increase in GDH activity (43%) ( Fig. 3B and C). On the contrary, when plants were grown under e[CO 2 ], the treatment with a mixture of N forms caused a slight increase in GS activity (19%) and a reduction in GDH activity (51%) ( Fig. 3A and C).

Discussion
The effects on plants of the N form(s) used for irrigation depend on several factors such as environmental conditions (temperature, light intensity, atmospheric CO 2 concentration, N concentration, average pH, and K supply), the proportions in which they are supplied (NO 3 − /NH 4 + ), the plant species, and even on the variety. Therefore, studies carried out by various authors about the use of different N sources have shown different results [14][15][16] .
Our data showed that the combination of different N forms (NO 3 − /NH 4 + ) and e[CO 2 ] provoked a significant increase in the gas exchange parameters. In our previous studies, we observed a similar behavior in pepper plants exposed to similar conditions 17,18 . Something similar was reported by authors such as Cruz et al. 19 and Bloom et al. 20 , who observed that plants (cassava and wheat plants, respectively) exposed to e[CO 2 ] and NH 4 + -based  . This suggests that futures increase in atmospheric CO 2 concentrations may compromise the productivity of some plants if we not change the fertilization strategies.
Curiously, other nutrients such as Ca 2+ and Mg 2+ increased in concentration with the 90/10 treatments at both CO 2 concentrations. However, authors such Boschiero et al. 23 reported contrary effects in sugarcane plants, which showed a reduction in the leaf nutrients with the increase in the NO 3 − /NH 4 + ratio. The fact that the 90/10 treatment produced an increase in the foliar Ca 2+ concentration, despite the antagonistic effect that exists between these two elements 18 , could indicate that at a low concentration of NH 4 + , this antagonistic effect does not occur for cucumber plants. A significant positive correlation was found between foliar Ca 2+ content and WUEi (R 2 = 0.6708, data no shown). It is known that Ca 2+ plays an important role in multiple photosynthetic pathways, it can both interfere with gas exchange by regulating stomatal movement, and can directly or indirectly regulate the activity of enzymes involved in photosynthesis. Authors such as Brestic et al. 24 observed that Ca 2+ improved Rubisco activity, and the higher activity seemed to be associated with a higher photosynthetic rate.
Concerning the values of K + , the finding that no differences were observed in foliar K + concentration between treatments, could be another indication that the NH 4 + concentration provided was not toxic for cucumber plants, as K + deficiencies have been observed in toxic concentrations due to competition in absorption between K + and NH 4 +23 . Regarding the CO 2 effect on the foliar nutrient content, curiously the cucumber plants responded in the opposite manner to the pepper plants under similar growth conditions 18 , which reinforces the idea of significant species-dependent-response.
Authors such as Teng et al. 25 , Markelz et al. 26 and Rubio-Asensio and Bloom 4 , observed that e[CO 2 ] provoked an increase in the photosynthesis rate of Arabidopsis, which was related with a rise in starch and the total nonstructural carbohydrates. In our experiment, something similar was observed with the starch, but no changes were observed in the soluble sugars. This could be due to the increase in starch or other carbohydrates storage polymers being greater than the increase in sugar concentrations, but the extent of the changes vary considerably between species 27 . Also, it is known that the N supplied can have an influence on the accumulation of starch and soluble sugars under e[CO 2 ] 27 . In our results, a relationship was once more observed between the increase caused by NH 4 + in photosynthesis and the higher leaf starch content observed. Regarding the behavior observed in the metabolic enzymes measured, our data showed that the treatment with NH 4 + under an atmosphere enriched with CO 2 provoked an increase in the GS activity and a reduction in the GDH activity, which indicate that the combination of the 90/10 treatment and e[CO 2 ] may promote the GS/ GOGAT pathway of N metabolism. This increase in the GS/GOGAT pathway could be partly responsible for a higher photosynthesis rate, as it would favor N assimilation 5 . On the contrary, the treatment with NH 4 + under a[CO 2 ] conditions resulted in a higher GDH activity and lesser GOGAT activity than the 100/0 treatment. These results suggest that GDH played a decisive role when the GS/GOGAT pathway was restricted. Authors such as Ma et al. 5 and Torralbo et al. 21 consider that the role of GDH in N metabolism becomes more important when plants are subjected to stress, specifically NH 4 + toxicity. GDH removes excess NH 4 + and thus reduces its toxic effect. To achieve a better understanding of the trends and relationships among all the studied parameters in relation to the N supply, a PCA was applied to the results. The results of the PCA are presented in Table 2 and presents a clearer distinction of the effects of nitrogen form and CO 2 concentration. The first two principal components (PCs) accounted for 71.38% of the total variance, attributing 42.26% to PC1 and 29.13% to PC2 ( Table 2). Most of the variables examined were positively correlated with PC1, and only two variables were negative correlated with PC1. The variables with the highest positive correlation coefficients were Ca 2+ (0.925) and WUEi (0.877), and others with a high correlation were soluble sugars (0.685), Mg 2+ (0.559), A CO2 (0.772) and NH 4 + (0.780). PC1 was negatively correlated with GDH (− 0.402) and K + (0.762), allowed for a clear separation of plants irrigated with NH 4 + in the nutrient solution, and suggested that plant growth with the 90/10 treatment was characterized by a higher Ca 2+ (Table 1), and higher WUEi (Fig. 1C). PC2 was positively correlated with GS (0.871), starch (0.690) and Ci (0.944), and was negatively correlated with GOGAT activity (− 0.669). The PC2 clearly separated plants grown under e[CO 2 ], characterized by a higher GS activity ( Fig. 2A), and higher Ci (Fig. 1B).
The data obtained in this experiment highlight the complexity and importance of using the correct type of nitrogen fertilization in the plant irrigation solution to face the environmental changes that are currently taking place (increase in CO 2 ). We have demonstrated that physiological parameters such as the A CO2 and WUEi can be improved in cucumber plants with the addition of NH 4 + in low amounts in the nutrient solution under a CO 2 -enriched atmosphere. Also, under these conditions (NH 4 + and e[CO 2 ]), the GS/GOGAT cycle is promoted, which favors the assimilation of N, and the increase in the concentrations of other nutrients such as NH 4 + , Ca 2+ , and Mg 2+ , and the starch content. www.nature.com/scientificreports/ Consequently, this study reveals the strong interaction between the N form supplied and e[CO 2 ], in terms of N assimilation, and therefore, of a better performance of the photosynthetic apparatus.

Material and methods
Plant material, growth conditions and treatments. Cucumber (Cucumis sativus L.), cv. Ashley seeds (Semillas Batlle, S.A., Barcelona, Spain) were germinated on a mixture of peat and perlite (3:1). Seedlings with two true leaves stages were selected for uniformity after the 12 days, and transplanted to 8-L black containers filled with coconut coir fiber (Pelemix, Alhama de Murcia, Murcia, Spain). Each container was rinsed with 1 L of water after transplanting. Irrigation was supplied by self-compensating drippers (2 lh −1 ), and fresh nutrient solution was applied with a minimum of 35% drainage.
The plant growth responses to different nitrogen forms and e[CO 2 ] were determined in an experiment carried out in a climate chamber designed by our department specifically for plant research proposals 28 , with fullycontrolled environmental conditions: 30% relative humidity, 16/8 h day/night photoperiod, an air temperature ranging from 28 to 20 °C, and a photosynthetically-active radiation (PAR) of 250 µmol m −2 s −1 provided by a combination of fluorescent lamps (TL-D Master reflex 830 and 840, Koninklijke Philips Electronics N.V., the Netherlands) and high-pressure sodium lamps (Son-T Agro, Philips). During the first seven days after transplanting (7 DAT), the plants were irrigated with Hoagland's solution (control), and then, the plants were irrigated with Hoagland's solutions that differed in their NO 3 − /NH 4 + ratios (in concentration percentages, 100/0 or 90/10) for twenty-two days.
The experiment lasted twenty-nine days and was carried out at standard CO 2 (400 µmol mol −1 CO 2 ) (a[CO 2 ]), and elevated CO 2 (1000 µmol mol −1 CO 2 ) (e[CO 2 ]) concentrations, with nine plants per treatment. Thus, four treatments were studied, corresponding to two nutrient solutions and two ambient CO 2 concentrations. Statistical analysis. Data were statistically analyzed using the SPSS 13.0 software package (IBM SPSS Statistics 25.0, Armonk, NY, USA), with an ANOVA and Duncan's multiple range test (P ≤ 0.05) using the treatments as a statistical variable to determine significant differences between means.
Gas exchange. The gas exchange measurements were performed just before starting the nitrogen treatments (7 DAT), and throughout the experiment (11,18,22,and 29 DAT). A CIRAS-2 (PP system, Amesbury, MA, USA) with a PLC6 (U) Automatic Universal Leaf Cuvette, was used to measure the net CO 2 assimilation (A CO2 ), internal CO 2 concentration (Ci) and instantaneous water-use efficiency (WUEi, A CO2 /E). The measurements were conducted on the youngest fully-expanded leaf from each plant. The cuvette provided light (LED) with a photon flux of 1300 µmol m −2 s −1 , 400 or 1000 µmol mol −1 CO 2 , 70% relative humidity, and a leaf temperature of 26 °C.
Ion concentrations. The NH 4 + , K + , Ca 2+ and Mg 2+ ions were extracted from ground leaf lyophilized (1 g) with bi-distilled water, and their concentrations were determined in an ion chromatograph (METROHM 861 Advanced Compact IC; METROHM 838 Advanced Sampler); the column used was a METROHM Metrosep C1 125/4.6 mm. www.nature.com/scientificreports/ Starch and soluble sugars. Soluble sugars were extracted by incubating 30-40 mg of lyophilized leaf tissue twice in 5 mL of 60% ethanol, 30 min each time, at 35 °C. The extract was centrifuged at 3500×g for 10 min at 20 °C, and the two supernatants were combined. Chloroform (5 mL) was added and the mixture shaken before centrifugation at 2700×g for 10 min at 20 °C. The sample was diluted fourfold with absolute ethanol to produce an extract in 80% ethanol for the measurement of soluble sugars according to Buysse and Merckx 29 . The residual material from the extraction with 60% ethanol was hydrolyzed with 3% HCl for 3 h at 125 °C, and the soluble sugars released were measured as an estimate of the starch content 30 .